Yaron, You are quite right about the source of the term 'source/sink', which I mixed up in my last post with rescue effect. I was off for a meeting when I wrote my post, and somehow switched tracks. Brown and Pulliam were both teachers of mine at U AZ, so I am especially embarrassed about the switch.
Your literature review is concise and useful. I would however argue that 'r > 0' is ambiguous. r can be seen as a temporally varying value with a mean that goes to zero at carrying capacity. Moreover it varies due to environmental stochasticity. Any healthy source population will have a _local_ r that is close to zero most of the time (barring stochasticity). The fundamental question is whether the population per capita growth rate is positive at low population densities. Hence it seems clearer to use the term 'r_m >0' for a source population. Patrick Foley [EMAIL PROTECTED]
